Table 1

Interview schedule

TDF domain	Questions
Knowledge	Evidence from the literature suggests that up to 22% of patients presenting for TKA will not have a clinically meaningful improvement from surgery. What do you think about this figure? How do you interpret the term ‘no clinically meaningful improvement’?
Knowledge	For the purposes of this interview, we are interpreting clinically meaningful improvement as no improvement in pain, function or QOL following surgery. Are you aware of what percentage of patients that you operate on do not benefit from surgery? How do you know this? Do you track it? Would you like to know? How could feed this information back to you? In what format?
Beliefs about capabilities	How confident are you in identifying patients who are unlikely to experience an improvement in symptoms from TKA? How good do you think you are at it compared with others? Do you feel you are unsure about identifying these patients at times? If so, what would you do?
Behavioural regulation	- Of all the patients referred to you, what is the percentage of patients that proceed to surgery and how many do you turn away? - What do you do with the ones that do not? Do you refer them somewhere?
Skills	What skills help you decide if someone is likely to benefit from surgery or not? Are you aware of any tools currently available to help you assess a patients’ risk of not responding? Do you use them? Why/why not?
Beliefs about consequences	Based on a set of evidence-based parameters, decision aids can predict the degree of risk that a patient will not achieve a clinically meaningful improvement from TKA. What do you think the benefits of using a decision aid might be? What might be the disadvantages of using an aid? Do you see anything legal or ethical about using a decision aid? Would the benefits outweigh the potential harms? Why?
Intentions, goal	Would using a decision aid influence your surgical decision making? Why/why not?
Reinforcement	What would motivate you to use a decision aid? Would you need to be presented with evidence from the literature? How would this evidence be best delivered? Who would it need to be delivered by?
Environmental context and resources	What would facilitate the use of a decision aid for you? How would it best be packaged? When do you think it would be best used? Do you think you are the best person to use it?
Decision process	If a decision aid predicted that patient had a 50 per cent risk of not benefiting from surgery, would you still operate? What about a 70 per cent? What would your level of acceptable risk be?
Social/professional role and identity	Do you think there would be agreement between surgeons on this cut point?
Social/emotional influences	What if something like this tool became compulsory – how would you feel? How do you think other surgeons would feel? How do you think patients would respond? Would their response influence your use of an aid? Would you worry about missing potential candidates who might have responded to surgery?
Optimism	How optimistic are you that a decision aid will reduce the rate of surgery in patients who are at high risk of not benefiting from surgery?

QOL, quality of life; TDF, Theoretical Domains Framework; TKA, total knee arthroplasty.